Return-Path: X-Original-To: apmail-flex-dev-archive@www.apache.org Delivered-To: apmail-flex-dev-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 2922D104B5 for ; Tue, 10 Dec 2013 18:22:47 +0000 (UTC) Received: (qmail 16089 invoked by uid 500); 10 Dec 2013 18:22:46 -0000 Delivered-To: apmail-flex-dev-archive@flex.apache.org Received: (qmail 16065 invoked by uid 500); 10 Dec 2013 18:22:46 -0000 Mailing-List: contact dev-help@flex.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@flex.apache.org Delivered-To: mailing list dev@flex.apache.org Received: (qmail 16057 invoked by uid 99); 10 Dec 2013 18:22:46 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 10 Dec 2013 18:22:46 +0000 X-ASF-Spam-Status: No, hits=1.7 required=5.0 tests=FREEMAIL_ENVFROM_END_DIGIT,HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of omuppi1@gmail.com designates 209.85.220.173 as permitted sender) Received: from [209.85.220.173] (HELO mail-vc0-f173.google.com) (209.85.220.173) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 10 Dec 2013 18:22:40 +0000 Received: by mail-vc0-f173.google.com with SMTP id ia6so4842868vcb.18 for ; Tue, 10 Dec 2013 10:22:19 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=iqW6lvhYbsKQcwnZPT64z58P8BMNQHN6u6XeomQc3Bs=; b=OfgFrX4yyKytwE56myv6AMIdggMJLaEVX4qIztGzpb/k27zqP4yRGC7M8Imn2XOoqy QdsnsHGvpfV4T7Lb2ZIy3XLesELNTTC9iBvSUl/AftVbBGnEUFvbxyPe/TJNsojWiNCO 9qBTgyFofausnBQ185DAp3CxYgscX25cHfgNGXoVV/mFBSBQN8lQ7N4knqcXhedHQ430 20Sa+0ycezqjgDQr/KjIba6uTOQ+boQ4JRiyotMZgIu4DBok5x60mBaiik61GBHy4VRD F9JBIRCStlwy3okXvrf0k/cDyW28qeB/ClHkRX/ZJlcqdjH6o01GjWZwB7qORts7kJ4W VxxQ== X-Received: by 10.220.224.134 with SMTP id io6mr397430vcb.74.1386699738943; Tue, 10 Dec 2013 10:22:18 -0800 (PST) MIME-Version: 1.0 Sender: omuppi1@gmail.com Received: by 10.58.206.17 with HTTP; Tue, 10 Dec 2013 10:21:48 -0800 (PST) In-Reply-To: References: <2095F5EBE04D59409DFCE91FFCEBF7AF3F5544DF@EXMBX05.netplexity.local> From: OmPrakash Muppirala Date: Tue, 10 Dec 2013 10:21:48 -0800 X-Google-Sender-Auth: EzPOvJIRjdwNRb2m43_C68t9-Ps Message-ID: Subject: Re: [Builds/Jenkins] Help and advise needed To: "dev@flex.apache.org" Content-Type: multipart/alternative; boundary=089e013a028659cbb704ed3231b4 X-Virus-Checked: Checked by ClamAV on apache.org --089e013a028659cbb704ed3231b4 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable On Tue, Dec 10, 2013 at 10:13 AM, OmPrakash Muppirala wrote: > On Tue, Dec 10, 2013 at 9:48 AM, Erik de Bruin wrote= : > >> I think he has the same MSDN as me, which should give him enough >> 'credits' to run a medium sized VM for ever... I currently run the >> Mustella VM as a large instance for about half a month, then switch it >> to medium for the remainder. A bit of a hassle, but worth it to run >> Mustella smoothly, I think. A 'regular' Jenkins server should run >> perfectly fine on a medium instance. >> >> EdB >> >> > Yes, that is correct. All Apache committers get a free MSDN subscription > [1] which is supposed to be used for the benefit of Apache projects. A > monthly credit for Windows Azure VMs comes with this subscription. > > My goal is to create a clone-able public image which has everything neede= d > to run Mustella builds that each committer can instantiate and deploy. I > want to make this whole process as simple as possible so that we can get = as > many committers' VMs up and running in no time. > > After that we will have one master Jenkins and set up all other VMs as > slaves. This way, we can first distribute VMs to do runs of various FP/A= IR > combinations. And then eventually, break down Mustella tests into variou= s > sections and have each slave run a subset, in order to reduce the time > taken for each Mustella run. > And we need to dedicate a VM specifically for building the SDK (that Erik > wanted) so as to reduce our dependence on builds@apache.org. > > I need a volunteer committer to help test out and document this process. > Any takers? > > Thanks, > Om > > [1] > https://svn.apache.org/repos/private/committers/donated-licenses/msdn-sub= scription.html > After the recent Mustella failures, I remembered that we need a VM for Alex's patch server as well. Thanks, Om > > >> >> >> On Tue, Dec 10, 2013 at 6:14 PM, Alex Harui wrote: >> > I don't think we actually know the cause of the problem. I am going t= o >> > continue to spend cycles to try to find out though. >> > >> > It would be nice to have an alternative to builds.a.o. I'm not sure i= f >> it >> > will cost Om money to run a builds server. >> > >> > -Alex >> > >> > On 12/10/13 2:01 AM, "Maurice Amsellem" >> > wrote: >> > >> >>I understand that. >> >> >> >>Actually, my "understanding" on this issue was that pixel bender >> compiler >> >>required some sort of hardware configuration (OpenGL, etc...) that wer= e >> >>not present on the b.a.o. new Windows Jenkins slave node, so that's wh= y >> >>the build was failing, and the Apache Infra was reluctant to let us >> >>modify anything, or even access the VM ourselves. >> >>So that's why I was proposing a "software only" solution. >> >> >> >>Now, it seems from what Om is saying that we can set up and use our ow= n >> >>Jenkins slave node VM. >> >> >> >>That, of course, is much preferable... >> >> >> >>Maurice >> >> >> >>-----Message d'origine----- >> >>De : Erik de Bruin [mailto:erik@ixsoftware.nl] >> >>Envoy=E9 : mardi 10 d=E9cembre 2013 10:44 >> >>=C0 : dev@flex.apache.org >> >>Objet : Re: [Builds/Jenkins] Help and advise needed >> >> >> >>Maurice, >> >> >> >>Your help is very much appreciated! >> >> >> >>I put "legal" in quotes, the issue is not really one of the law, more = of >> >>the rules. An Apache release is supposed to be 'source only', and we i= f >> >>we can produce needed binaries from source, we keep only the source, n= ot >> >>the artefacts themselves in the repo. >> >> >> >> >> >>EdB >> >> >> >> >> >> >> >>On Tue, Dec 10, 2013 at 10:34 AM, Maurice Amsellem >> >> wrote: >> >>>>In addition to the various "legal" issues with binaries in the repo. >> >>> I understand it's not a good idea to have binaries in the repo, so I >> >>>won't insist. >> >>> But please can you explain what are the legal issues of having >> binaries >> >>>in the repo? Is this because of Adobe, or ASF rules ? >> >>> >> >>> On a side note, I was just trying to help, with my limited >> >>> understanding and knowledge, and because the email thread was titled >> >>> "help and advise needed" ;-) >> >>> >> >>> Regards, >> >>> >> >>> Maurice >> >>> >> >>> -----Message d'origine----- >> >>> De : Erik de Bruin [mailto:erik@ixsoftware.nl] Envoy=E9 : mardi 10 >> >>> d=E9cembre 2013 09:12 =C0 : dev@flex.apache.org Objet : Re: >> >>> [Builds/Jenkins] Help and advise needed >> >>> >> >>> In addition to the various "legal" issues with binaries in the repo, >> >>>we'd be masking the cause of this failure. In order to prevent furthe= r >> >>>deterioration of the build process, we need to figure out what went >> >>>wrong and fix it. >> >>> >> >>> EdB >> >>> >> >>> PS. Thanks for leaving the keyboard on the Mustella VM set to FR... >> >>> Took me while to figure out that I hadn't gone insane or if my >> >>> keyboard was broken ;-) >> >>> >> >>> >> >>> >> >>> On Tue, Dec 10, 2013 at 9:04 AM, Maurice Amsellem >> >>> wrote: >> >>>>> someone who does use the pbj's will grab the nightly and complain. >> >>>> >> >>>> I don't understand. >> >>>> Why would someone complain if the pbj's are in the nightly? >> >>>> >> >>>> [From the other emai] >> >>>>>Apache repos aren't supposed to contain compiled code. The pbj fil= es >> >>>>>were removed during the initial release audit. >> >>>>>I don't think a workaround can involve checking in the pbj files. >> But >> >>>>>we could borrow them from a prior release package temporarily. >> >>>> >> >>>>>So we could make the compilation conditional on a env parameter, a= nd >> >>>>>set that in the Jenkins job accordingly? >> >>>>>Yes but ... >> >>>> >> >>>> Alex, the conversation is getting out synch, so I am not sure that = I >> >>>>have understood what you said. >> >>>> >> >>>> So can we include the pbj in the repo, and have a parameter to >> >>>>conditionally compile the pbj ? >> >>>> - This parameter would be set by default to do the compilation (so >> >>>> that folks can recompile) >> >>>> - and turned off on the b.a.o vm, with pre-compiled pbj's. >> >>>> >> >>>> >> >>>> Maurice >> >>>> >> >>> >> >>> >> >>> >> >>> -- >> >>> Ix Multimedia Software >> >>> >> >>> Jan Luykenstraat 27 >> >>> 3521 VB Utrecht >> >>> >> >>> T. 06-51952295 >> >>> I. www.ixsoftware.nl >> >> >> >> >> >> >> >>-- >> >>Ix Multimedia Software >> >> >> >>Jan Luykenstraat 27 >> >>3521 VB Utrecht >> >> >> >>T. 06-51952295 >> >>I. www.ixsoftware.nl >> > >> >> >> >> -- >> Ix Multimedia Software >> >> Jan Luykenstraat 27 >> 3521 VB Utrecht >> >> T. 06-51952295 >> I. www.ixsoftware.nl >> > > --089e013a028659cbb704ed3231b4--